How many lumens should be cultured in the conservative diagnosis of catheter-related bloodstream infections?

Abstract

Recent practice guidelines for the diagnosis of catheter-related bloodstream infection (CRBSI) describe as an "unresolved issue" the number of lumens from which blood culture specimens should be drawn to make a conservative diagnosis of CRBSI. Our objective was to determine how many CRBSI episodes would be missed if not all catheter lumens were sampled. We performed a retrospective study (1 January 2003-31 May 2009) in patients with microbiologically proven CRBSI in which all available catheter lumens (those that did not contain clots) were used to draw blood culture samples. We calculated the number of episodes that would have been missed in double- and triple-lumen catheters if the culture of samples obtained from 1 lumens had been eliminated. We studied 171 episodes of proven CRBSI in 154 patients. Overall, if 1 lumen-associated culture had been eliminated for both double-lumen and triple-lumen catheters, we would have missed 27.2% and 15.8% of episodes of CRBSI, respectively. If we had eliminated 2 cultures for triple-lumen catheters, 37.3% of episodes would have been missed. Samples for blood culture should be obtained through all catheter lumens to establish a diagnosis of CRBSI.